Gertrude-Emilia Costin, Ph.D., M.B.A. Institute for In Vitro Sciences, Inc. (IIVS)

NorCal SOT Fall Symposium 2017

The 3Rs

28 September 2017, San Francisco, CA, USA

Modernizing the “six-pack” testing strategy: influx of modern in vitro techniques

Consumer/ End-user

Safety

Industry/ Manufacturer

Safety/ Testing Labs

Labeling/ Regulatory

Agency

Perspectives, challenges, common goals and working together

Presentation Outline

Current regulatory climate – global acceptance of in vitro methods

The reductionist concept of in vitro methods

Drivers of in vitro methods advancement

Beyond the “six-pack” battery of acute toxicity tests

Acute oral toxicity

Acute dermal toxicity (oral vs dermal route comparison)

Acute inhalation toxicity

Ocular irritation (the EPA OPP testing strategy)

Skin irritation/corrosion

Skin sensitization

Modernizing the “six-pack” testing strategy: influx of modern in vitro techniques

Current regulatory climate – global acceptance of in vitro methods

The reductionist concept of in vitro models

G.-E. Costin and H. A. Raabe. In vitro toxicology models. In: The Role of the Study Director in Non-clinical Studies. Pharmaceuticals, Chemicals, Medical Devices, and Pesticides. (Eds. William Brock, Barbara Mounho and Lijie Fu), John Wiley and Sons (2014). G.-E. Costin. Advances in science: next generation of lab tools, models and testing platforms used in predictive toxicology. Molecular Life 2017; 1(1), 22-28, doi: 10.26600/MolLife.1.1.3.2017. Available at: http://molecular-life.org/wp-content/uploads/2017/07/Advances-science-next-generation-lab-tools-models- testing-platforms-used-predictive-toxicology.pdf.

“Less is more”

Whole animal (Rabbit)

Organ - Eyeball (Enucleated chicken or

rabbit eye)

Tissue - Cornea (Resected bovine

cornea)

Cell culture (Statens Seruminstitut Rabbit cornea cells)

1940s 1990s

Tissue construct (Human EpiCorneal™

model)

Organ-on-a-chip (Human retina)

Body-on-a-chip (Human organotypic

microtissues)

2000s 2010s

Cell culture (Normal human

corneal epithelial cells)

Retinal Pigment Epithelium

Semi-permeable Membrane

Vascular network Nutrient channels

Fibrinogen- endothelial cell administration channel

Drivers of in vitro methods advancement

Industry

CROs

Trade Associations

Animal Welfare Groups

Academia

Regulatory Agencies

Public

In vitro methods

Ongoing evolution on so many levels • Improve scientific basis for testing using human-derived

test models • Reduce the number of animals for testing • Increase predictivity • Reduce time, price • Harmonize requirements and prediction models http://ntp.niehs.nih.gov/pubhealth/evalatm/test-method-evaluations/project-milestones/index.html#atc-inh. http://alttox.org/mapp/table-of-validated-and-accepted-alternative-methods/.

R

Refinement

Beyond the “six-pack” battery of acute toxicity tests

PESTICIDES

Acute oral rat

Acute dermal rabbit

Acute inhalation

rat

Skin sensitization guinea pig

mouse

Ocular irritation

rabbit

Skin irritation

rabbit

EPA Health Effects Test Guidelines OPPTS 870.1000 Acute Toxicity Testing-Background.

The modern in vitro toxicology perspective

Acute oral toxicity – CRO’s perspective

Treatment termination

NRU addition

Dilution/Dosing Cell seeding NHEK 3T3

Pre-testing: Solubility

assessment Solvent addition

Plate reading

Test system: normal human keratinocytes (NHEKs) or Balbc 3T3 mouse fibroblasts Assay endpoints: cell viability [by Neutral Red Uptake (NRU) assay] Data calculation: estimated log LD50 mmol/kg = 0.435 x log mean NRU50 mM + 0.625 For initial dose setting OECD Guidance Document 129

Efforts to use a single route

• 2013, US EPA OPP, National Toxicology Program (NTP) Interagency Center for the Evaluation of Alternative Toxicological Methods (NICEATM): retrospective analysis of oral and dermal acute lethality studies focused on formulated pesticides products considering the EPA pesticide categorization scheme.

Acute dermal toxicity – Regulatory perspective (oral vs dermal route comparison)

Acute dermal toxicity – Regulatory perspective (oral vs dermal route comparison)

Creton et al. Acute toxicity testing of chemicals – Opportunities to avoid redundant testing and use alternative approaches. Crit. Rev. Toxicol. 40: 50-83 (2010). Seidle et al. Examining the regulatory value of multi-route mammalian acute systemic toxicity studies. ALTEX 28, 2/11 (2011). Moore et al. Can acute dermal systemic toxicity tests be replaced with oral tests? Regul. Toxicol. Pharmacol. 66: 30-37 (2013).

Data analysis • For 57% of the 592 formulations, the

results of both oral and dermal acute toxicity studies fall within the same Toxicity Category.

• For 38% of the formulations, the oral study falls within a lower (i.e., more protective) Toxicity Category.

• Thus, for 95% of the formulations in the analysis, if the dermal study had not been available and labelling had been based only on the Toxicity Category for the oral acute toxicity study, the PPE requirements on the labelling would have been equally protective or more protective.

• For the remaining 5%, factors other than the dermal acute toxicity may influence PPE labeling requirements.

The dataset of rat acute oral and acute dermal LD50 studies included: • 592 formulated pesticide products, representing

316 active ingredients • all four Toxicity Categories • 13 different formulated pesticide product types

The agency has used this analysis to support a policy statement in Section 3.0 to waive all acute lethality dermal studies for formulated

pesticide products.

Acute inhalation toxicity – CRO’s perspective Efforts for in vitro methods development

Thorne D, Adamson J. A review of in vitro cigarette smoke exposure systems. Exp Toxicol Pathol 65, 1183-1193 (2013). http://www.mattek.com http://www.epithelix.com *Image courtesy of Dr. Khalid Amin, University of Minnesota, Department of Laboratory Medicine and Pathology

Smoke exposure systems Cell lines/tissues/explants • Bronchial epithelial cells • Reconstructed tissues

EpiAirway model (MatTek Corporation)

Normal, human-derived tracheal/bronchial epithelial cells

MucilAir model (Epithelix) Primary epithelial cells co-cultured with human airway

fibroblasts

Endpoints • Cytotoxicity • mRNA (MUC5AC) • Protein expression: IL-6, IL-8, MMP-1 • Gene expression and

microarrays • Cellular glutathione

levels

Precision Cut Lung Slice (PCLS) Alveolar spaces; H&E staining,

40x magnification*

https://www.epa.gov/sites/production/files/2017-04/documents/session-3-2st-century-toxicology-update.pdf. http://www.piscltd.org.uk/wp-content/uploads/2016/09/Acute-Inhalation-Workshop_2016_ALowit.pdf. https://www.epa.gov/pesticides/epa-releases-guidance-voluntary-pilot-program-reduce-animal-testing.

Voluntary pilot program underway where registrants may send the in vivo acute lethality study for oral and inhalation formulation/product testing as currently required and simultaneously submit the calculations using the GHS dose additive mixtures equation.

Acute inhalation toxicity – Regulatory perspective

Ocular irritation - the EPA OPP testing strategy A continuum of sensitivity

*Clorox *Colgate Palmolive *Dial *EcoLabs *JohnsonDiversey (currently SealedAir) *P&G *SC Johnson *The Accord Group *Institute for In Vitro Sciences (IIVS)

Extreme Severe Moderate Mild Very Mild

Color Cosmetics

Cleaning Products

Industrial Chemicals

EPA I EPA II EPA III EPA IV

GHS 1 GHS 2 GHS Non-classified

Multiple in vitro assays

Multiple in vitro assays

Household

Currently, one in vitro assay is not sufficient for all eye irritation categories – therefore a bottom-up/top-down strategy was proposed

Antimicrobial Cleaning Products (AMCP) Labeling Requirements

AMCPs quick facts

• Contain ~275 different active ingredients. • Are marketed as sprays, liquids, concentrated

powders, and gases. • More than 5000 are currently registered with

the U.S. Environmental Protection Agency (EPA) and sold in the marketplace.

http://www.epa.gov/opp00001/factsheets/antimic.htm.

The vast majority of the household and commercial cleaning products do not have to go through a registration process before they are marketed. Companies decide how to assure safety – generally without using animals (in vitro).

Antimicrobial claim

The product is now EPA regulated (animal testing for safety is required).

Both EPA and industry agreed to work together to build a predictive and conservative in vitro strategy designed to replace the requirement for Draize rabbit eye irritation data with non-animal methods.

BOVINE CORNEAL OPACITY AND PERMEABILITY (BCOP) ASSAY Test system: Viable corneas maintained in culture Assay endpoints: opacity and permeability Data calculation: In Vitro Score = Opacity + (15 x Fluor OD490) OECD TG 437 US EPA OPP policy (3-2-2015): Use of an alternate testing framework for classification of eye irritation potential of EPA pesticide products - 40CFR Part 158W for AMCPs (anti-microbial cleaning products)

Fluorescein Addition

Corneal Excision

Mounting Initial

Opacity Test Substance

Exposure Rinsing Permeability

Endpoint Fixing the Corneas

Ocular irritation - the EPA OPP testing strategy

PREDICTIVITY • Only 2 of 61 materials (8%)

were under-predicted. • All of the EPA toxicity Category

IV materials are over-predicted as Category III since the BCOP does not seem to be able to differentiate between materials at this lower end of the toxicity scale.

LIMITATIONS • If the anti-microbial cleaning product

is a High Solvent (>5 solvent) formulation, it should be tested in the BCOP assay using a 3 minute exposure instead of the normal 10 minute exposure.

• Testing of ketones and alcohols in the BCOP has been shown to result in high false positive rates for the assay, but not all ketones or alcohols are over-predicted.

LABELING APPLICABILITY • The BCOP assay does

differentiate between EPA Category I and II materials, so it is most useful in this higher range.

BCOP Assay Overall Performance

Receipt of tissues

Treatment of tissues

Rinsing of tissues and MTT reduction

Isopropanol extraction of MTT

Reading of plates

3D EPIOCULAR™ (EO) ASSAY Test system: Human three dimensional (3D) reconstructed tissue model (keratinocytes) Assay endpoints: tissue viability Data calculation: the exposure time required to reach a 50% reduction in tissue viability (ET50 value) dependent on cytotoxic potential and rate of penetration US EPA OPP policy (3-2-2015): Use of an alternate testing framework for classification of eye irritation potential of EPA pesticide products - 40CFR Part 158W for AMCPs

Ocular irritation - the EPA OPP testing strategy

EpiOcular™ Assay Overall Performance PREDICTIVITY

• There was only one under-prediction for the 41 total materials.

• The EO method was able to

clearly separate a few EPA Category IV materials, although most Category IV materials will be over-predicted as Category III.

LIMITATIONS • Oxidizing materials should not

be tested in the EO, but both water soluble and water insoluble materials can be tested.

LABELING APPLICABILITY • The EO assay should be

useful in clearly identifying materials as EPA Category III or Category IV, but cannot separate EPA toxicity Category I from Category II.

CYTOSENSOR MICROPHYSIOMETER (CM) ASSAY Test system: Mouse fibroblasts (L929) Assay endpoints: real-time measurement of cellular metabolism Data calculation: dose calculated to reduce the population metabolic rate to 50% of the initial rate (MRD50) OECD TG: draft (2012) US EPA OPP policy (3-2-2015): Use of an alternate testing framework for classification of eye irritation potential of EPA pesticide products - 40CFR Part 158W for AMCPs

Seeding Overnight incubation

Transfer into cytosensor

chamber Transfer into

the CM Stabilization

Serial dilution of the test

substance Dosing cycle

(20 min) Calculation of metabolic rate

Ocular irritation - the EPA OPP testing strategy

Cytosensor Assay Overall Performance PREDICTIVITY

• There were no under-predictions of EPA toxicity categories (of a total of 108 cleaning products tested).

• 89% of the Category IV materials

were over-predicted as Category III or higher. However, the CM was able to clearly identify some Category IV materials.

LIMITATIONS • Oxidizing materials, or

materials not completely aqueous soluble at the highest dilution, should not be tested in the CM.

LABELING APPLICABILITY • The CM should be useful in clearly

identifying materials as EPA Category III or Category IV, but cannot separate EPA toxicity Category I from Category II.

BC

OP

In

Vit

ro

Sc

ore

BCOP scores vs. EPA Category (Draize) - example

< 4 min

Evaluate components

Oxidizing chemistry?

Expected severe or

moderate?

Water soluble?

BCOP

In vitro score

CM

In vitro score

In vitro score

No No

Yes Yes

> 25 <75

<2 mg/ml

>80 mg/ml

>2 but < 80 mg/ml >4 but < 70 min

>70 min

Category I

Category II

Category III

Category IV

Default Category I; To distinguish Category II

from I, conduct BCOP

EO

No Yes

<25

Ocular irritation Outline of the in vitro testing strategy

Use of an alternate testing framework for classification of eye irritation potential of EPA pesticide products. U.S. EPA (2015).

BCOP = Bovine Corneal Opacity and Permeability CM = Cytosensor Microphysiometer EO = EpiOcular™

Default Category III; To distinguish Category IV from III, conduct CM or EO

>75

Program implementation

In 2009, the US EPA instituted a 18 month Pilot Program in which manufacturers of AMCP submitted data using the proposed in vitro testing strategy for ocular irritation in place of animal testing for product registration.

The program became permanent in 2013. The policy was updated in March 2015.

Selected reasons for success

• Limited applicability domain (AMCP) • Cooperation among companies provided a larger test substances set • Continued interactive discussions with EPA’s Office of Pesticide

Programs • Animal test variability highlighted • Purpose of test (hazard labeling, not preclinical) clearly understood • Approach proposed was very conservative (few false negatives, but

many over-predictions)

Skin irritation/corrosion

Receipt of tissues Treatment of

tissues Rinsing of tissues and

MTT reduction Isopropanol

extraction of MTT Reading of

plates

3D RhE ASSAY Test system: Reconstructed human epidermis (RhE) tissue model (keratinocytes) Assay endpoints: tissue viability Data calculation: % viability OECD TGs: 431 (corrosion; updated 2016); 439 [Skin Irritation Test (SIT); updated 2015]

Prediction Model: Skin Irritation

In vitro result Mean tissue

viability) In vivo

prediction UN GHS

CATEGORY

≤ 50% Irritant (I) Category 2

> 50% Non-irritant (NI) No Category

Desprez B., Barroso J., Griesinger C., Kandarova H., Alepee N., Fuchs H.W., Two novel prediction models improve prediction of skin corrosive sub-categories by test methods of OECD, Toxicology in Vitro, 29, 2055-2080 (2015) .

Prediction Model: Skin Corrosion

EpiDerm™ (EPI-200); SkinEthic™ RHE; epiCS®

STEP 1 < 50% after 3-minutes exposure Corrosive ≥ 50% after 3-minutes exposure AND < 15% after 60-minutes exposure Corrosive

≥ 50% after 3-minutes exposure AND ≥ 15% after 60-minutes exposure Non-corrosive

STEP 2

<25%; 18%; 15% after 3-minutes exposure Corrosive: Optional Sub-category 1A

≥25%; 18%; 15% after 3-minutes exposure Corrosive: A combination of optional Sub-categories 1B-and-1C

Viability measured after exposure time points (3, 60 and 240-minutes)

Prediction to be considered UN GHS Category

EpiSkin™ (SM)

< 35% after 3-minutes exposure Corrosive: Optional Sub-category 1A

≥ 35% after 3-minutes exposure AND < 35% after 60-minutes exposure OR ≥ 35% after 60-minutes exposure AND < 35% after 240-minutes exposure

Corrosive: A combination of optional Sub-categories 1B-and-1C

≥ 35% after 240-minutes exposure Non-corrosive

Skin irritation: US EPA registration of products Decision process using the rabbit Draize skin irritation test

Chemical hazard classification and labeling

Primary Dermal Irritation Index (PDII)

Sum erythema (1/24/48/72 hr) + Sum oedema (1/24/48/72 hr) 4 intervals (1/24/48/72 hr) x no. of animals

US EPA Hazard Category

I II III IV

PDII Corrosive

>5.0 2.1-5.0 0-2 Irritation Potential

Severely Irritating

Moderately Irritating

Slightly Irritating

Signal Word DANGER WARNING CAUTION CAUTION

http://www2.epa.gov/sites/production/files/2014-07/documents/chapter7_revised_final_0714.pdf.

OECD TG 431

OECD TG 431

2

1.5 2.3

5

4

US EPA Hazard Categories US EPA PDII

II > 5

III 2.1-5.0

IV 0-2

UN GHS Hazard Categories

≥2.3≤4.0 2

≥1.5<2.3 3

<1.5 Not Classified

UN GHS Reaction Scores

Skin irritation: US EPA registration of products

Assessment of an alternative approach using OECD validated in vitro assays

OECD Guideline for the testing of chemicals No. 431, In vitro skin corrosion: human skin model test (2016). OECD Guideline for the testing of chemicals No. 439, In vitro skin irritation reconstructed human epidermis test method (2015).

OECD TG 439 ?

Corrosive Severe Moderate Mild to Non-irritant Draize irritation scale

OECD TG 439

Tissue Receipt (EpiDerm™)

Tissue Treatment

Tissue Rinsing

Post-Treatment Incubation

MTT Reduction

Isopropanol Extraction

Spectro photometric

Quantification

Exposure Time (EX)

Post-Treatment

(PT)

Assay Designation

15 min 42 hr EX15/PT42

24 hr EX15/PT24

60 min 42 hr EX60/PT42

Optimization of the validated in vitro Skin Irritation Test (SIT)

Ass

ay M

odifi

catio

n (T

issu

e Tr

eatm

ent)

EX: 60 min.

EX: 15 min.

Retrospective Data

Ass

ay M

odifi

catio

n (P

ost-T

reat

men

t Inc

ubat

ion)

PT: 42 hr

PT: 24 hr

New Data

Clorox Ecolab Procter & Gamble Reckitt Benckiser S. C. Johnson & Son SealedAir

US EPA Category determined

in vivo

US EPA Category determined in vitro

II III IV

Sensitivity 100% 46.2% 92%

Category under predicted

0 23.0% NA

Category over predicted

NA 30.8% 8.0%

Performance of the proposed US EPA Prediction Model

EX15/PT42 ≥ 20%

EX60/PT42 Category II

Category IV

Category III

EX15/PT24

< 20% > 20% ≤ 20%

> 20% and

if EX15/PT24

> 20% and

if

Proposed in vitro testing strategy for assignment of US EPA hazard categories for skin irritation

Mechanisms of skin sensitization

EPIDERMIS

DERMIS

LYMPH NODE

Chemical (hapten) penetrates the skin and reacts with protein(s)

Chemical is recognised by Langerhans cells (LC) which then migrate from the skin to the draining lymph node

Mature LC presents chemical to T cells

This causes proliferation of specific T cells

Increased number of chemical-specific T-cells released into the systemic circulation

Inflammation

Subsequent skin contact with chemical activates the T cells and leads to clinical manifestation

INDUCTION

ELICITATION

Courtesy of Dr. D. Basketter

Adverse Outcome Pathway (AOP) for skin sensitization

Human Cell Line Activation Test (monocyte cell line THP-1)

Direct Peptide Reactivity Test

Organ Response

Proliferation of T-cells in lymph

nodes

Organism Response

Dermal inflammation (after challenge)

Penetration into the viable

epidermis

Molecular properties

INDUCTION ELICITATION Cellular

Response

Expression of cell surface markers and cytokines

LC activation h-CLAT

Keratinocyte activation KeratinoSens

LuSens

Molecular Initiating Event

Electrophilic reactivity

Covalent interaction with proteins

Peptide reactivity DPRA

GPMT LLNA HIRPT

AOP

phas

e

In silico In chemico In vitro In vivo Clinical

QSAR

Test

m

etho

d

Bi

olog

ical

ev

ent

Evans, CC and Fleming, JD. Allergic contact dermatitis from a henna tattoo. N. Engl. J. Med. 359: 627 (2008).

Skin sensitization – CRO’s perspective DIRECT PEPTIDE REACTIVITY (DPRA) ASSAY Test system: in chemico Assay endpoints: HPLC determination of peptide depletion OECD TG 442C

Mean of Cysteine % Depletion

Reactivity Prediction

0% - 13.89% Minimal Non-sensitizer 13.89% - 23.09% Low Sensitizer 23.09% - 98.24% Moderate Sensitizer 98.24% - 100% High Sensitizer

Mean of Cysteine and Lysine % Depletion

Reactivity Prediction

0% - 6.38% Minimal Non-sensitizer 6.38% - 22.62% Low Sensitizer

22.62% - 42.47% Moderate Sensitizer 42.47% - 100% High Sensitizer

Prediction Model

Sample preparation Separation Module Data analysis

Peak Area of un-reacted peptide is compared to peak area of reacted peptide

Minutes

Un-reacted Peptide

Depleted Peptide

Skin sensitization – CRO’s perspective

Pre-testing: solubility assessment

Cell dosing

Treatment termination

Addition of luciferase

Addition of MTT

Sensitization endpoint

Cytotoxicity endpoint

KERATINOSENS ASSAY Test system: HaCaT cells (immortalized keratinocytes containing a reporter construct with a copy of the Antioxidant Response Element (ARE) of the human AKRIC2 gene upstream of a luciferase gene Assay endpoints: induction of luciferase activity, cytotoxicity Data calculation: EC1.5 value (test substance concentration for induction 1.5 fold time above threshold) Imax (the largest average gene fold induction above 1.5 by the test substance) Cimax(the test substance concentration at which the largest average fold induction value is achieved) OECD TG 442D

A test substance will be considered to have sensitization potential if:

1) The EC1.5 value falls below 1000 µM (or 200 µg/mL) in at least 2 of 3 repetitions;

2) At the lowest concentration with a gene induction above 1.5, cellular viability should be greater than 70%

3) An apparent overall dose response should be similar between repetitions.

Prediction Model

Chemical allergen

CD54 CD86

YYYYYYY APC anti-CD 54

YYYYYYY PE-Cy7 anti-CD86

Y Y

Sensitizer

CD54/APC + CD86/PE-CY7 -

Human Cell Line Activation Test (hCLAT) CRO’s perspective

CD54/APC + CD86/PE-CY7 +

CD54/APC - CD86/PE-CY7 +

CD54/APC - CD86/PE-CY7 -

CD54>200% CD86>150%

Prediction Model

hCLAT ASSAY Test system: human monocytic leukemia cell line (THP-1 cells) Assay endpoints: CD86 and CD54 cell surface market expression Data calculation: Relative Fluorescence Intensity (RFI) using flow cytometry Prediction model: CD86 ≥150% and CD54 ≥200 with cell viability of ≥50% in at least 2 independent repetitions relative to vehicle controls OECD TG 442F

Human Cell Line Activation Test (hCLAT) Regulatory perspective

International Cooperation on Alternative Test Methods (ICATM)

• First International Cooperation on Alternative Test Methods (ICATM)

Workshop was held in 2016 and was focused on the international regulatory applicability and acceptance of alternative non-animal approaches. The countries participating were USA, EU, Japan, Korea, Canada, Brazil and China.

• Multiple non-animal testing strategies incorporating in vitro, in chemico,

and in silico inputs demonstrate comparable or superior performance to the LLNA.

• A planned product of the ICATM workshop is the development of an

assessment framework for integrated non-animal approaches that could serve as replacements for the current animal test, the LLNA for multiple chemical sectors (pesticides, cosmetics, pharmaceuticals, industrial chemicals, etc.)

https://www.epa.gov/sites/production/files/2017-04/documents/session-3-2st-century-toxicology-update.pdf.

Modernizing the “six-pack” testing strategy: influx of modern in vitro techniques

PESTICIDES

Acute oral rat Acute

dermal rabbit

Acute inhalation

rat Skin

sensitization guinea pig

mouse

Ocular irritation

rabbit

Skin irritation

rabbit

OECD 431

OECD 442C

OECD 437

EPA OPP

R

CROs

Industry

Trade Associations

Animal Welfare Groups

Academia

Regulatory Agencies

Public

OECD 129

Perspectives, challenges, common goals and working together

Trade Associations

Animal Welfare Groups

Academia

Industry/ Manufacturer

Labeling/ Regulatory

Agency

Safety/ Testing Labs

Validated

Transferable

Specific

Relevant

High-throughput Sensitive

Reliable

Reproducible

Easy to perform Affordable

Consumer/ End-user

Safety

Acknowledgments

IIVS Team Devin Sheehan Victoria Diersen Rebecca Pham Asha Pidathala Megan Lamm Lindsay Krawiec Elizabeth Sly Hans Raabe Rodger Curren

Industry Consortium Clorox Colgate Palmolive Dial EcoLabs SealedAir P&G Reckitt Benckiser SC Johnson The Accord Group

US EPA OPP Jennifer Mclain Anna Lowit

The Accord Group Pat Quinn